INHERITANCE STUDIES IN SOYBEANS. 111. SEED-COAT ...INHERITANCE STUDIES IN SOYBEANS. 111. SEED-COAT COLOR AND SUMMARY OF ALL OTHER MENDELIAN CHARACTERS THUS FAR REPORTED' F. V. OWEN

INHERITANCE STUDIES I N SOYBEANS. 111. SEED-COAT COLOR AND SUMMARY OF ALL OTHER

MENDELIAN CHARACTERS THUS FAR REPORTED'

F. V. OWEN Maine A ~ ~ i c ~ l t u r a l Experiment Station, Orono, Maine

Received February 8 , 1927

TABLE OF CONTENTS

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Seed-coat color types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pigments responsible for seed-coat colors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The variability of seed-coat colors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results of previous investigatio . . . . . . . . . . Allelomorphic series for restrict

Restriction of pigment to the hi1 t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Restriction of pigment to form an eyebrow pattern. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Relation between black and brown pigment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Flower and pubescence colors in relation to seed-coat pigments. . . . . . . . . . . . . . . . . . . . . . . . . .

Linkage versm independence of seven pairs of factors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Restriction of pigment over the en

Duplicate factors for tawny pubescence. . . . . . . . . . . . . . . . . . . . . . . . .

Linkages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

PAGE

50 51 52 53 54 56 58 59 62 62 63 68 69 72 72 73 74 75

INTRODUCTION

In 1923 the writer began a study of the inheritance of seed-coat colors in the soybean (Soju wax). From a practical standpoint, the problem is of great interest because seed-coat colors are very important in the selection of pure strains and in the detection of mixtures in commercial varieties. This led to a rather intensive study of the factors concerned in mottling, the results of which are being published in another paper.

The results of previous workers indicated that two varieties may look alike but may behave very differently in crosses. This made it advisable to secure numerous crosses among a large number of varieties in order that the genetic constitutions of different types could be compared. Al-

1 From the Department of Genetics Paper No. 74 and the Department of Agronomy, WIS- CONSIN AGRICULTURAL EXPERIMENT STATION. Published with the approval of the Director of the Station.

GENETICS 13: 50 Ja 1928

SEED-COAT COLOR IN SOYBEANS 51

though the original plans have by no means been realized, over 300 hybrids, representing 35 different combinations, have been produced.

Some of the parental plants used for hybridization were from well known varieties, but others were from selections made by the Agronomy Depart- ment of the WISCONSIN AGRICULTURAL EXPERIMENT STATION, and their origin is uncertain. It has been necessary to assign arbitrary numbers to these selections which have not been identified as belonging to any particular variety.

With but few exceptions all factors studied in soybeans have something to do with the color of the seed coat, either directly or through linkage. It seems logical, therefore, to center the discussion on seed-coat colors although a complete list is given of all the genetic factors that have been reported.

In carrying out the problem the writer has been greatly aided b y counsel from Professor B. D. LEITH of the Agronomy Department and Professors L. J. COLE and R. A. BRINK of the Genetics Department of the UNI- VERSITY OF WISCONSIN. It is a pleasure to acknowledge all the advice and constructive criticisms which have been received.

SEED-COAT COLOR TYPES

NAGAI (192 1) and PIPER and MORSE (1923) have given classifications of seed-coat color types for soybeans, but for the purpose of interpreting Mendelian characters the following classification has been found most useful.

I. Self-color type 1. 2.

3. 4.

Black Brown (may be of different intensities according to the variety and somewhat according to conditions at the time of maturity) Imperfect black Buff

} (Associated with gray pubescence)

11. Bicolor type 1. Black mottling pattern on brown background

1. Black 2. Brown 3. Imperfect black 4. Buff

IV. Green or yellow seed coat with dark hilum. (Mottling is developed over the remaining portion of the seed coat according to environmental conditions.)

111. Eyebrow pattern with green or yellow background

(Associated with gray pubescence) 1 GENETICS 13: Ja 1928

52 F. V. OWEN

1. Hilum black 2. Hilum brown 3’ 4. Hilumbuff

imperfect (Associated with gray pubescence)

V. Green or yellow seed coat with light hilum. (Mottling is produced according to environmental conditions but pigments are always potentially present .) 1. Mottled with black 2. Mottled with brown 3 . Mottled with imperfect black 4. Mottled with buff 1 cence)

(Associated with gray pubes-

PIGMENTS RESPONSIBLE FOR SEED-COAT COLORS

What has been designated as black pigment in the seed coat of the soybean is really a very intense purple and belongs to the general class of anthocyanins. The brown or phlobaphene-like pigments are undoubtedly closely related to quercetin but their exact chemical composition is not known. In the imperfect blacks there is a mixture of the two pigments and even in the self-blacks a considerable amount of brown pigment is masked by the black.

In addition to the pigments mentioned above, which are glucosides and have many properties in common, there is another blackish-brown pigment which behaves very differently. In all probability it belongs to the general class known as “oxidation pigments” because of their connection with the process of oxidation. This is the pigment responsible for the common blackish-brown pod color in black-podded varieties but this same pigment may also find its way into the parenchymatous cells of the seed- coat. Frequently it produces such a smudgy appearance that a classification of a plant in regard to the other pigments is rather uncertain. Unlike the anthocyanins and phlobaphene-like pigments, the production of the oxidation pigment may be greatly increased by injury to the pod and testa.

Another colored condition has been described by MATSUMOTO and TOMOYASU (1925) which they have called “purple speck.” The pigment responsible for this abnormal coloration, according to the descriptions, has similar properties to anthocyanins but a fungus (Cercosporina Kiku- chii) was found to be responsible for its occurrence.

The black and brown pigments, with which the present article is chiefly concerned, are independent of the green and yellow plastid pigments. In self-browns or mottled types not all of the plastid color is concealed,

SEED-COAT COLOR I N SOYBEANS 53

so it is usually possible to detect the presence of green when it is present but in self-blacks plastid colors are entirely masked.

THE VARIABILITY O F SEED-COAT COLORS

Except in self-colored types there is a great deal of variability in the amount of pigment produced in the seed coat. This has been regarded as objectionable by practical growers, so the subject of mottling has received considerable attention in all the important soybean-growing sections of the United States. Much effort has been spent in selecting for non-mottled strains but all varieties which have been tested in cooperation with the Agronomy Department of the WISCONSIN AGRICULTURAL EXPERIMENT STATION have produced mottling in some degree. Some selections have proved to be much more badly affected than others but mottling seems to be a general phenomenon with all yellow- and green-seeded varieties. Details of this work are being published in another paper.

WOODWORTH and COLE (1924) have given evidence to show that environmental factors may be important in bringing out mottling pigments and their conclusions have been well confirmed during the progress of this study. Varieties which failed to develop any signs of mottling in the greenhouse during winter months were badly mottled in the field.

A rich loam soil has consistently produced much more mottling than a light sandy soil, but the distance of spacing plants has also been found to be very important. Plants grown three feet apart have produced more mottled seed than plants grown closely together, but on certain types of soil the distance of spacing has not been a determining factor. These results are confirmed by HOLLOWELL’S (1924) findings a t Iowa.

Although it seems probable that there may be many factors which affect mottling, perhaps the most important is the maturation of the seed. An extreme amount of mottling was found to be associated with plants which remained green after the normal time for the seeds to mature.

Environmental factors have no apparent influence in the case of self- colored varieties, because no difficulty is experienced in getting these varieties to breed true; but there are noticeable differences in intensity of color due to environment. The imperfect-black type is extremely variable in its appearance. Sometimes seeds are nearly black and in other cases some experience is required to distinguish this type from light brown. NAGAI (1924) gives evidence to show that this variability is not independent of inheritance but the present writer has no data on the relative importance of environment and heredity on this particular coloration.

GENETICS 13: Ja 1928

54 F. V. OWEN

The self-browns and buffs are also variable in regard to the intensity of color and conditions a t maturity seem important in this connection.

For a genetic study of seed-coat colors it is usually desirable to have the plants growing in an environment that will produce the maximum amount of pigment. This together with an intimate knowledge of the variability that might be expected due to environment is necessary in order to classify intelligently certain Fz progenies.

RESULTS OF PREVIOUS INVESTIGATIONS

PIPER and MORSE (1923) found black seed coat dominant to brown. In the particular example cited a single factor was responsible for the difference. NAGAI (1921) also found black to be dominant to brown but he distinguished between different types of blacks and browns. He demon- strated the presence of a factor R which produced anthocyanin pigment and a factor C which increased the intensity of pigmentation. According to NAGAI’S hypothesis, with CR present the seed coat is black; with cR it is an imperfect black; with Cr it is brown; and with cr it is a very light brown or “buff.”

In addition to C and R, NAGAI assumed that another factor 0 must be present in order to explain the inheritance of certain browns. A rich reddish brown was recessive to the common dull brown. Therefore, with CO present the pigment was a dull brown, and with CO, reddish brown; but the relation between the two factors, C and 0, was not entirely clear.

WOODWORTH (1921) observed that two factors, B and H , were necessary for the production of black pigment in the hilum, but H was found to be completely linked to T , the factor for tawny pubescence. Plants homozygous for ht always produced gray pubescence and a seed coat with a brown hilum. In Part I1 of this series another factor, E , was added to this linkage group. This factor E is a partially dominant factor for early maturity but since the expression was not always clear-cut it was only through the linkage relation that its effect could be attributed to a single factor. There was evidently a small percentage of crossing over but this could not be accurately measured.

NAGAI and SAITO (1923) found self-blacks or self-browns recessive to yellow or green and they have described three restriction factors. Accord- ing to these writers, the factor K inhibits the production of pigments forming a “patched” or “eyebrow” pattern; I inhibits nearly completely the formation of chromogenic substance; and H inhibits it entirely. The recessives of these factors have no effect.


NAGAI and SAITO also found a single factor M responsible for the production of a black mottling pattern upon a brown background. They assumed that M is linked to the factor P which inhibits the formation of pubescence, giving a crossover value of approximately 18 percent. In Part I1 of this series the writer also reports a linkage between P and RI, a factor for anthocyanin pigment. This linkage appears to be nearly complete with only occasional crossovers. Mention is also made of the factor p 2 for glabrousness reported by STEWART and WENTZ (1926) which was recessive to P 2 , a factor for pubescence. Thus we have a character (glabrousness) which is dominant in one case and recessive in another case; apparently due to different factors.

TAKAHASHI and FUKUYAMA (1919) have given results obtained in F2 from five different crosses but no attempt was made to give an interpretation, due to the small amount of data available. Their results a t least indicate that varieties are green or yellow because of the presence of factors that inhibit pigment formation. More blacks than browns were secured in the F2’s but the data do not warrant conclusions in regard to the factorial relationships.

NAGAI (1921) found that black and brown pigments are formed regard- less of whether the plastid pigments in the seed coat are green or yellow. Green, however, was dominant over yellow. WOODWORTH (1921) obtained identical results but found that V , the factor used for green seed coat, was linked with D, a factor for the production of yellow cotyledons, giving about 13 percent crossing over; but V was assumed to be independent of another factor I which is a duplicate of D and also produced yellow cotyledons. These findings have been well confirmed by the writer and the results are briefly presented in Part I of this series.

TERAO (1918) found a single factor responsible for green seed coat color when the maternal parent was yellow, but when the maternal parent was green there was no segregation, the Fl’s and F2’s being the same color as the female parent. PIPER and MORSE (1923) confirm TERAO’S findings, and both writers have postulated that two types of green are responsible for the odd results. Although the inheritance of the green seed coat was independent of the paternal parent yet two types of greens were distinguished in the F2 progenies of green seed-coat o Xyellow seed-coat 3. One type was more intense than the other and when counts were made it was found that the results were approximately in accord with a ratio of 3 dark green to 1 light green. One type of green was therefore inherited according to Mendelian expectation while the other type was a clear case of maternal inheritance. GENETICS 13: Ja 1928

56 F. V. Om’EX

ALLELOMORPHIC SERIES FOR THE RESTRICTION O F PIGMENT I N THE SEED COAT

NAGAI and SAITO (1923) postulated that the three restriction factors which they described as H, I, and K, were allelomorphic. This assumption has been well confirmed in the present study, so there seems to be no good reason for not making this clear in the terminology. Four allelomorphs may, therefore, be described according to their inhibiting effect upon the production of pigment in the seed coat. The following figure illustrates the action of the factors which have been found in this allelomorphic series.

L ight Dark Eyebrow Self - hilum hilum p a t t e r n color

As the above figure suggests Ih, under the proper environmental conditions, inhibits all pigment formation in the seed coat; Ii inhibits pigments to the hilum; Ik inhibits the formation of pigment in such a way as to form the “eyebrow” or “patched” pattern; and i, the recessive allelomorph, has no effect.

Although it is necessary to define these restriction factors rather definitely, it should be clearly kept in mind that environmental conditions unfavorable for pigment formation are necessary in order for these restriction factors to fully express themselves. Varieties with the genetic constitution i i come under Class I or I1 of the “Seed-coat Color Types.” In these two classes there is no restriction of the pigment and environment has practically no effect but in classes 111, IV, and V, which IkIk, IiIi, and IhIh represent respectively, environmental conditions are very important in addition to the hereditary factors. Thus, varieties which come in Class V are of constitution IhIh, but may develop considerable mottling. From individual seeds it would frequently be impossible to distinguish the three types because of the variation due to environment but when all the seeds from a plant are available for classification there is usually no difficulty. It should be mentioned, however, that these restriction factors are only partially dominant, A cross between IhIh and i i invariably


produces an F1 that is mottled more than plants homozygous for Ih and in such cases it is sometimes difficult to distinguish the color types.

Unfortunately the writer was not successful in obtaining a cross between varieties of constitution i i and I k I k but NAGAI and SAITO (1923) have already described this relationship. All other combinations, however, have been made and the results from these crosses constitute a critical test for the theory that Ih, Ii, I k , and i are allelomorphic. The best evidence for this relationship rests upon the results of the three combinations of crosses between Mandarins ( IhIh) , Soysotas (i i) and Man- chus ( I%I%). The parental plants of the Mandarin and Soysota varieties were from strains selected which were known to be homozygous, and the Manchu selection (No. 59), which will be referred to here, was also known to be homozygous.

The distribution of dominant and recessive types in all Fz progenies closely approached a 3 : 1 ratio and the crucial proof for the allelomorphic series rests upon this simple but consistent behavior. The following tabulation indicates this relationship.

Mandarin (ZAP) X Soyosota (i i) 15 5 O.Ok1.3 Manchu (PZi) XSoyosota (i i) 250 92 6.5k5.4 Mandarin (ZAZA) X Manchu (ZiZS)I 223 75 2.0,s .O

Dominant Recaaire Denialion

Any explanation other than that of an allelomorphic series is obviously precluded because this is the only way in which a 3 : 1 ratio can be secured for each of the combinations. If the Mandarin variety were of constitution HHi i and if the Manchu were hhI1 ( H being synonymous with Ih and I synonymous with I % ) the crosses with the Soysota variety could be explained because it may have the constitution hhii. Such a theory is shattered when the Mandarin is crossed with the Manchu because with two factors involved there would be a dihybrid ratio and 1/16 of the plants in Fz would be self-colored, but none were observed. The probability of this happening when 412 plants are grown would be (15/16)412 or the odds would be over 10l1:1, and results from other similar crosses are also against a two factor difference.

The following crosses were also obtained in which the factor I k is involved. In the cross Black Eyebrow ( P I k ) XGlabrous No. 5 ( I z I%) , 186 plants produced seed with the pigment restricted to the hilum and 65 produced seed having the eyebrow pattern. Based on a 3: 1 ratio the

There were 412 F1 plants in this cross but only tawny-pubescent plants were classified for reasons which will follow in the text.


58 F. V. OWEN

deviation is 2.3 f4.6. If I k were not an allelomorph of Ii then the Glabrous No. 5 parent would necessarily have had the two dominant factors I i and I k . Such an assumption is not in accord with the breeding facts from crosses between varieties with pigment restricted and self-color because in every case a single factor difference explains the results.

In order to furnish further proof for the allelomorphic series it would have been desirable to have backcrossed an F1 plant of constitution IhIi or I h I k to the self-colored i i variety. Such crosses should bring out only the two dominant types in the next generation, and the proof for the allelomorphic series would be complete. Due to the difficulty in crossing soybeans, however, this task would be difficult to carry out and the indirect evidence which is already available is practically conclusive that Ih , Ii, Ik, and i occupy identical positions on the chromosomes.

Restriction of pigment to the hilum

It is impossible to assert definitely that a factor identical with NAGAI and SAITO’S factor I has been found, but a t least it is convenient to use such a designation. This factor li, as it is being defined here, is responsible for restricting pigments to the hilum.

The Manchu variety has a yellow seed coat with a black hilum and results have shown that the reason for the seed coat being yellow instead of black is due to the presence of .Ti, the dominant restriction factor. Environmental factors produce great variations in the amount of mottling developed; but the hilum remains colored in spite of the fact that environmental conditions may be very unfavorable for the development of mottling.

With no restriction factors other than Ii present the hilum is always black or brown, but the intensity of the color may depend on other factors. In tawny-pubescent varieties the hilum color is prominent but in gray-pubescent varieties the color may be very faint. Varieties differ also in regard to size of hilum and this likewise seems to determine the prominence of the color. Of course, we can not be absolutely sure from the data now available that the factor responsible for restricting pigments to the hilum has been the same one in every case, but neither have we proof for the contrary. I t seems logical, therefore, to assume that Ii is responsible for restricting pigments to the hilum.

Three different strains of Manchus were crossed with self-browns but each of these combinations gave very similar results. Table 4 gives the detailed behavior of these crosses but the combined results obtained in F2 are the following:


P he n d r w a Black hilum Brown hilum Selfabck S e l f d r " RIP r S Rli r6

Observed 584 169 171 71 Calculated 559.7 186.6 186.6 62 .2

Ditference +24.3 -17.6 -15.6 +8.8 - -- - -

The value of x2 is 5.26 assuming independence between IC, the restriction factor, and R 1 the factor for blaek pigment. This gives P a value of 0.16.

Another cross already mentioned was made between the Black Eyebrow variety (IkIk) and a glabrous Japanese variety (1212) (No. JS) with a brown hilum. The Black Eyebrow variety is characterized by the eyebrow pattern that has been described so we assume that the restriction factor I k is present. The factor It has the same action with the eyebrow pattern as with the self-color, inhibiting the development of pigment outside of the hilum in both cases. The results are as follows:

Phtrpca Black hilum Brown hilum Black qw3row Brown evebrow RiP rtP R,P ' r@

Observed 136 50 47 18 Calculated 141.2 47 .1 47.1 15.7

Ditference -5.2 +2.9 -0.1 +2.3 - - - -

It may be safely assumed that Ii and R1 were independent because of the small deviations from the theoretical 9:3:3:1 ratio and the value for x2 is 0.329 and P =0.9+.

Restriction of pigment over the entire seed coat NAGAI and SAITO (1923) have described a factor H which completely

inhibits the development of brown and black pigments. They neglect to state that mottling is possible when H is present but there are many reasons to believe that such is the case. The writer's experience has been that the expression of this factor, I h according to the new usage, may be greatly influenced by environmental conditions. All pigment may be entirely restricted or there may be considerable mottling. The Mandarin, Aksarben, and Ito San varieties have such restriction factors and these three varieties have been crossed rather extensively.

When a self-color (brown or black) is crossed with a variety possessing the restriction factor Ih , the F1 has a strong tendency to be intermediate. Even in the greenhouse, where conditions extremely unfavorable for the production of mottling were provided, heterozygous plants produced seeds slightly speckled or blotched with pigment. This condition was not apparent, however, when the cross was made with the Manchu variety GENETICS 13: Ja 1928

60 F. V. OWEN

which possessed the factor Ii for the restriction of pigment to the hilum. It is to be assumed, therefore, that with both Ii and Ih present the tendency for a restriction of pigment is stronger than if only one Ih is present.

In order to have a convenient method of referring to pigment patterns the designations used in the classification of color types have been adhered to. “Dark hilum,” therefore, means that the hilum is colored as in the Manchu variety but the pigment may be black or brown. The factor I’ is responsible for inhibiting the pigment to the hilum in this manner. “Light hilum” means that I h is present and no pigment is formed except that which may be due to mottling.

No difficulty was experienced in classifying tawny-pubescent plants according to whether their seeds had dark or light hilums, but with gray- pubescent plants the distinction has not always been clear because of the faint development of pigment. In some way the presence of T seems to intensify seed-coat pigments and this helps materially in classifying plants for the presence or absence of I h .

Four different combinations between self-browns and yellows have been made. The detailed results are given in Table 5 but the combined results are the following :

Phenotypes Tawny moltled Tawny self-bruwn Gray mottled Gray, Buff T l h T i f P t i

Observed 137 48 38 22 Calculated 137.8 45.9 45.9 15 .3

Difference -0.8 +2.1 -7.9 +6.7

A comparison with calculated results, on the basis of independence, gives deviations that may be accounted for as being due to chance; the value of x2 is 4.39 and P=O.23.

Six combinations between black-seeded varieties and varieties possessing the factor Ih were also secured (table 1). The combined Fz data from these crosses are as follows:

Mottled or colorless Selj-colored Den P . E . Dec. P. E . -

518 186 10.0 7.75 1.3

These data again show a fairly good agreement between observed figures and the theoretical 3: 1 ratio.

A cross, Black Eyebrow (IkIk)XIto San (IkIh) , was also made which gave an actual ratio of 18 with pigment restricted to 4 with the eyebrow pattern. These data are almost in perfect agreement with theoretical figures, the deviation being 1.5 k 1.4.


LIQET EILW

Black I Brown

A cross between a Manchu No. 1 ( I i I i ) and an unidentified yellow variety ( P I h ) , designated as No. 12 in the records, gave the following results in Fz.

DARK EILIJY

Black -1 Brown

Observed Calculated

Difference

120 36 51 14 124.4 41.4 41.4 13.8

- 4 . 4 - 5 . 4 +9 .6 +0.2 -- __I ___ -

Comparing the observed data with the calculated, assuming that I h is independent of RI, x2 = 3.10 and P = 0.38. The behavior of FI progenies also corroborates the results obtained in F2 and these data are given in table 6.

Six combinations between light hilum and gray pubescence ( t I h ) Xdark hilum and tawny pubescence (T Ii) were secured. From a general ob- servation of these Fz’s one would naturally assume that I i is completely linked with T , because no plants with gray pubescence bore seed with the typical dark hilum. If complete linkage is assumed we have the following results compared with the theoretical 2 : 1 : 1 ratio.

Tawny, Light hilum Tawny, Dark hilum Grab. Light hilum Gray, Dark hilum

Observed 709 235 299 Calculated 621.5 310.7 310.7

Difference $87.5 -75.7 -11.7

With this comparison x2=31.20 and P=O.OOOOOl, so it is unlikely that the deviations could be due to chance. Since T , the factor for tawny pubescence, has a marked effect on both the quantity and quality of pigments, one may assume that some gray-pubescent plants really had dark hilums but could not be easily distinguished. Recalculating with this assumption we get the following comparison with the 9 : 3 : 4 ratio.

Tawny, 1,ignt hilum Tawny, Dark hilum Gray, (Hilum colm n d die- tinguiehed)

Observed 709 235 299 Calculated 699.2 233.1 310.7

Difference +9.8 + 1 . 9 - 1 1 . 7

In this case x2=0.59 and P=0.9, showing a very good agreement between observed and expected results.


62 F. V. OWEN

A very similar cross between the dark and light hilums has been described by WOODWORTH (1921) but he was unable to classify his Fz plants, because of the irregular pigment which was probably due to environmental conditions.

Restriction of pigment to f o r m an eyebrow pattern

Although the three restriction factors Ih , Ii, Ik seem well established, there are some facts which indicate a larger series of allelomorphs. It has already been mentioned that some strains of Manchus have been selected which mottle much more badly than others, but selection within any one strain has not been successful. In other words these strains must have been homozygous to begin with, and if mutations occur they have not been with sufficient frequency to be selected. In progeny tests with selections from the Black Eyebrow variety, however, heritable variations from the normal eyebrow pattern have been observed, indicating the presence of another factor which restricts the pigment less than Ik, but it is possible that modifying factors were involved.

One selection (No. 84) bred true for the normal eyebrow pattern for four generations and the uniformity was very remarkable. Another selection (No. 82) bred true for a very dark pattern, and in this progeny plants with half of their seeds completely black were not uncommon, but no plant produced 100 percent black seeds. A third progeny (No. 83) was apparently heterozygous for factors affecting the eyebrow pattern because selections were made which bred true for a normal eyebrow pattern; others bred true for a very dark pattern; and still other selections con- tinued to produce some plants with normal patterns and others with dark patterns.

RELATION BETWEEN BLACK AND BROWN PIGMENT

It has been pointed out that self-black is dominant to self-brown in the seed coat and that ratios of 3 blacks to 1 brown are secured in FS. The first cross made by the writer was of this type and the following results were obtained.

Bhck Brown Des. P. E. 72 28 3 2.92

In this cross (Black, No. 3XBrown, No. 30)’ the F1 plant bore seed fully as black as the maternal parent, and in the Fz’s there was no visible

1 This progeny also segregated for an unusual type of hilum peculiar to the Soysota variety, the paternal parent. Ordinarily there is a clean separation between the hilum and the funiculus in soybeans, but in this abnormal type the Malpighian cells are arranged in such a way that the funiculus is torn, leaving a rough surface that is very characteristic. This condition is recessive


difference between the intensity of blacks, that is, black was fully dominant to brown.

In 1923 a natural hybrid (Selection No. 60) was accidentally found that segregated according to a ratio of 3 blacks to 1 brown in the progeny that was grown, but all heterozygous plants bore seed slightly speckled with brown. The results were as follows:

Observed Calculated

Black Dilute black Brown 7 14 8 7.25 14.50 7.25

It so happened that the observed distribution was exactly in accord with the calculated except for an extra brown. Results in Fa proved that all of the 14 plants with dilute black were actually heterozygous. The peculiar behavior of the black pigment is noticeable only in the heterozygous condition. In homozygous plants the black is apparently just as intense as that produced by any other variety. It is, therefore, a simple case of incomplete dominance. This kind of partial dominance is well known, and 1 : 2 : 1 ratios are perhaps as much to be expected as 3 : 1 ratios. BLAKESLEE and AVERY (1917) have reported a case of incomplete dominance of mottling in the Adzuki bean (said to be Phaseolus Murcgo) that is very similar to this lack of dominance in the seed-coat color of the soybean.

The difference between the two kinds of blacks appeared interesting, so crosses were made using the hybrid Fz plants for parental material. Crosses with types possessing the dominant inhibiting factor I h were secured. Actually this was equivalent to back-crossing because in some cases heterozygous plants were used as parents. Therefore, combinations with rl were secured as well as with R1.

Two combinations with R1 are given in table 1, Mandarin and Aksarben varieties being the other parents. Without considering the details of the table a t this time it may be of interest to state that out of 17 self-blacks grown in 1925 from these crosses 10 were dilute black and were undoubtedly heterozygous, as was previously found to be the case.

FLOWER AND PUBESCENCE COLORS IN RELATION TO SEED-COAT PIGMENTS

In table 1 six crosses are compared between blacks with a factor present for the production of anthocyanin pigment and yellows with I h present for the restriction of all pigment. It has already been pointed out that

to the normal hilum and in all crosses it has proved to be independent of all other perceptible characters. Tables 9 and 10 give the evidence for this statement, the factor for normal hilum being designated as N and that of the Soysota variety as B.

GENETICS 13: Js 1928

64 F. V. OWEN

blackxbrown has usually given 3: 1 ratios, black being dominant. WOOD- WORTH’S (192 1) example, where complementary factors were involved giving a 9: 7 ratio, is the only exception reported.

In the crosses reported in table 1 three strains of blacks (No. 3, No. 29,‘ and No. 60) and four strains of yellows (Mandarins, Ito Sans, Aksarbens, and No. C3) were used. No. C3 is an unidentified Chinese variety, and it resembles the Mandarin variety very much but the flowers are white. It may also be said that this is the only white-flowered variety used in the crosses referred to in table 1. The other six parents had the ordinary purple flowers which are dominant to white.

Black No. 60 was a selection from the natural hybrid that has been described, the peculiar thing about this selection being the incomplete dominance of the black seed-coat color in the heterozygous condition.

Table 1 affords an opportunity for a study of pubescence colors in relation to seed-coat colors. Ito Sans have tawny pubescence while the other three yellow-seeded varieties have gray pubescence. When Ito Sans were crossed with blacks a 9:3:3:1 ratio resulted in Fz and the results can be explained due to the presence of RI in the blacks and I h in the Ito Sans, but these results are quite different from those obtained when gray-pubescent varieties were used as parental stock. When Mandarins and the Aksarbens were crossed with blacks a 27 : 9 : 9: 3 : 9: 3 : 3 : 1 ratio was found to explain the results. I t will be noted that rather high values for x2 are obtained for these crosses in table 1 but the numbers are rather small and the fact that all classes are represented may be of some sig- nificance. Combining progenies, 133,139,172,175, 171, and 164 x2 = 14.52 and P =0.04. Both the black and the brown pigments were influenced in some way by the presence of T. On plants with gray pubescence, of constitution tt, there was an incomplete development of pigment. The black and brown correspond with what NAGAI (1921) has described as “imperfect black” and “buff ,” respectively. It is therefore convenient to continue the use of these terms.

NAGAI does not state that imperfect black was associated with gray pubescence but this has been the case in progenies 133, 172, and 164 (table 1). The logical assumption, therefore, is to consider T identical with NAGAI’S factor C. If these factors are not identical they are a t least completely linked, because tawny pubescence has always been associated with a more intense development of both the anthocyanin and phlobaphene-like pigments. These are the exact properties which NAGAI has attributed to his factor C.

1 Selection from Wisconsin Black variety. Nos. 3 and 60 could not be identified as to variety.

SEED-COAT COLOR IN SOYBEANS

Black

Iottled Self-color

65

Brown

Mottled Self-coloi __-_____

--_.---

1 3 5 4 0 1 3 4 4 1

TABLE 1 Fz progenies of blackXyellow seed coats.

Buff -- vlottled --

--

--

9 . 8 __.-

-0.8 --

11 6 . 5

+4.5

lo

-- --

14 6 . 9

+7.1 -- --

71 5.9

+1.1 __.-

-

OQENI K B E l

Self-mI4 -

-

__ 7 1

2 2 9 1

3 .

-2 .

4 0

7 3 3 2.

-0.

2

4 2 4 2 .

+l .

41 2.

+2

26 34.4

-8.3

72

103 88.6

+1 -1 ----__c_

18 11 6 11.4 11.4 3.6 __---_

+6.6 -0.4 +2.i __---_.L_

21 18 10

3 1 7 6 7 28 24 17 29.5 29 .5 9.2

-14.4

30

26 56 58.2

--

-2.2

31

25 56 62.4

-6.4

70 7 0 . 3

-0.3

-1.5 -5.5 +7.:

11 19 3

6 10 2 17 29 5 19.4 19.4 6.1

- -- -_

I____--

-2.4 +9.6 -1.i ------

13 14 4

11 10 0 24 24 4 20 .8 20.8 6.1 ----

f 3 . 2 +3.2 -2. ' --_L-___

25 23.4

--- +1.6

B U T PVBEBCBINCE

Imperfect black

Mottle --

-- lo

8 18 29.

Black (No. 3) XIto San

Calculated

Difference

150 y2= 1.25

P= 0.74

Black (No. 29) XIto San

Calculated

Difference -

Black (No. 29) X Mandarin

Cdcdated

Difference --

Black (No. 60) X Mandarin

Calculated

DiEerence

127

- 139

133 rota]

- 175

172 rota

- 171

164 rota

y2= 7.24

P = 0.07 _c_-

y2 = 14.89

P = 0.04

x2 = 12.56

P= 0.08

-- 7

3 10 9.

-11.

6

5 11 19.

-

-8.

9 --

8 17 20,

+O.

3

3 6 6 .

-

-0.

3 -

2 5 6 .

Black (No. 60) XAksarben (No

53)

Cdculated

Difference

x2 = 12 .64

P= 0.08

x2= 3.21

P= 0.67

-3 -1. -

42 5.

Black (No. 3) XYellow

(No. C3) Cdculoted

Difference

142 1.51 17,

-2 -1.

Plants also had white flowers. All plants had purple flowers. Progenies 139 and 171 are reciprocals of 133 and 164, respectively.


Progeny No. 142 [black No. 3 (rlRzTIV)Xyellow No. C3 (r,rztw)] is of special interest because no browns appeared in the tawny-pubescent class, but in the gray-pubescent class there were both imperfect blacks and buffs. This gives a ratio of 15:l for the presence versus absence of anthocyanin pigment. In the gray-pubescent class the imperfect blacks and buffs were very similar to seed similarly classified in the progenies 133, 172, and 164. F3 progenies were also grown from Fz plants of progeny 142. The original hypothesis was confirmed by these results. Although only 10 progenies from plants of phenotypic constitution (T RZIh) were grown yet every possible genotypic arrangement was secured. Two segregated the same as the previous F2 generation, five segregated for I h but not for T, and three segregated for T but not for Ih . In the later progenies, Rz was linked completely with T and there were no signs of crossing over.

It is of special interest to note that the imperfect blacks in progeny 142 were also associated with purple flowers and that buffs were associated with white flowers. Therefore, it seems entirely probable that W, the factor for purple flowers (WOODWORTH 1923), may also be responsible for the development oi anthocyanin pigment in the seed coat in this particular case. To obtain further evidence FS progenies of phenotypic constitution T Rsiw were grown in 1926 and two of these proved to be heterozygous for T. The linkage between T and Ra was again complete and all gray-pubescent plants produced buff seed-coat color. This evidence corroborates the previous breeding behavior in showing the necessity of the presence of W before anthocyanin pigment can be produced in the seed coat of gray-pubescent plants.

WOODWORTH found that purple flower color was associated with purple hypocotyls but no effect on the seed coat has been reported; yet this condition is not surprising because in other plants the development of anthocyanin pigment has often been found to be associated with flower color. COMBES (1912) working with a species of Spiraea, found that coloration in the bark of twigs was easily brought about by decortication on species with colored flowers but this was not possible with species having white flowers.

As already stated, the crosses Black x Mandarin or Black X Aksarben can readily be explained by the three factors, RI, T, and I h . The restriction factor I h is independent of the other two factors, so there is no need of considering it a t this time. The expression of the other two factors is as follows:

&T. . . . . . . . . . . . . . . . . . . .Black, tawny R,t .................... .Imperfect black, gray r lT. . . . . . . . . . . . . . . . . . . . .Brown, tawny r l t . . . . . . . . . . . . . . . . . . . . . .Buff, gray


In the cross, Black (No. 3)XYellow (No. C3), the situation is also simple but there is difficulty in explaining why the behavior is so different from other crosses. It cannot be said that Rl is responsible for the production of anthocyanin in this case because of the complete linkage. Per- haps we can postulate that another factor Rz, not allelomorphic to RI , is the factor for anthocyanin formation and that it is completely linked with T. This factor may be identical with Woodworth’s factor H because H also produced black pigment and was completely linked with T . The factor W has no apparent effect when T is present but when the constitution was tt it produced an imperfect black. The question then arises- why is it possible that W can have this effect? If it had produced the same effect in Mandarins and Aksarbens the seed coats on these varieties should have been mottled with imperfect black instead of light brown because these are purple-flowered varieties. The genetic constitution of the yellow selection No. C3 must, therefore, differ from the genetic constitution of Mandarins in such a way that when W is present a certain amount of brown pigment is converted into anthocyanin. Perhaps r2, an allelomorph to R?, makes this possible; and if this is the case a cross between the Mandarin (with W ) and selection No. C3 (with r z ) should give imperfect blacks in F1; i.e., the complementary factors W and r2 are necessary to explain the results.

On the basis of the above assumption the black selection No. 3 (rlR2TW) must have a factor Rz which is different from the factor R1 in No. 29 and No. 60, and selection No. C3 (rlrztw) must also have an additional factor r2 that is not present in Mandarins and Aksarbens. Then the results in progeny 142 can be explained as follows:

.Black, tawny T,RJW 1 . . . . . . . . . . . . rlR2T-d f ” “ rlrztW. . . . . . . . . . . . . . . . . .Imperfect black, gray rlrztw. . . . . . . . . . . . . . . . . . .Buff, gray

Since r2 was completely linked with t , the single factor T may be assumed to be responsible for the production of black seed coats as well as tawny pubescence, but results from other crosses show that T does not always have this effect. When selection No. 3 (rlR2T) was crossed with the Ito San (rlra’T) there were plants with brown seed coats and tawny pubescence in Fz (table 1) . Furthermore, rz’ of the Ito San variety must be considered different from rz of selection No. C3 because r2 was a complement of W for the production of anthocyanin in the imperfect black seed-coats of progeny 142 (table l) , but rz’ does not have this effect. From


68 F. V. OWES

these results it seems possible that an allelomorphic series may be situated at the 1 locus. If this is true the following combinations have now been identified :

R2T. . . . . . . .Tawny pubescence, black seed-coat (selection No. 3) rd . . . . . . . . . .Gray pubescence, seed-coat imperfect black with W but buff with M (selection

*’IT.. . . . . . . .Tawny pubescence. seed-coat brown with either W or M (It0 San variety) I’J. . . . . . . . Gray pubescence, seed-coat brown with W or w (Mandarin and Aksarben varieties).

No. C3).

The identity of the factor Rz for the production of anthocyanin pigment in the seed coat is well established, but it has been found in only one selection (No. 3 ) . In selection Nos. 29 and 60 the factor (RI) responsible for the production of anthocyanin pigment in the seed coat is clearly distinct from Rz, because, unlike the latter, it is not linked with T . There is also good reason for using the designation RI in selection No. C4 (table 3 ) but a number of other combinations were made between varieties both of which had tawny pubescence. In these crosses the factor for the production of anthocyanin may have been either RI or Rz but for convenience RI has been used in case of doubt.

DUPLICATE FACTORS FOR TAWNY PUBESCENCE

The assumptions that have just been made depend upon T being the same factor for tawny pubescence in each case. It might have been assumed that there is only one factor R for black seed-coat color and two factors for tawny pubescence. A convenient designation would be TI and T,. The factor TI could be independent of R, but R and T z would be completely linked, Such an assumption would explain the results as well as the method which has been chosen.

Further work is necessary to decide these relationships but the results from one Fz population give a real indication for the presence of two factors for tawny pubescence. This was progeny No. 71 (No. Sp 3-9X No. 29) which has been discussed under Part I of this series, as evidence for maternal inheritance of cotyledon color. Both paternal parents were homozygous for tawny pubescence but there was segregation in Fz giving 71 tawny to 15 gray pubescent plants. Assuming duplicate factors the deviation is 10.6+4.8. It is very evident that this is a rather wide deviation from a 15: 1 ratio but since the odds are only about 6: 1 it does not seem entirely improbable. Fs progenies from this cross are to be grown and results reported at a future date.


LINKAGE versus INDEPENDENCE O F SEVEN PAIRS OF FACTORS

Eleven crosses between Mandarins and an unidentified Chinese variety, given the arbitrary number C4, proved successful. These varieties differ in seven pairs of characters. These contrasted characters and the factors which correspond to them are as follows:

Mandarin

Purple flower color, W Yellow pods, I Gray pubescence, t Yellow seed-coat, g Light hilum, I h Brown mottling pigment, 11 Yellow cotyledon color, DlDz

Green No. C4

White flower color, w Black pods, L Tawny pubescence, T Green seed coat, G Dark hilum, I i Black mottling pigment, RI Green cotyledon color, 4 d Z

The factor pairs Gg, IhIi, and Rlrl have the same meaning here as in NAGAI and SAITO’S (1923) descriptions; and Ww, Ll, Tt, and Dldl, have been described by WOODWORTH (1921 and 1923). The factor D2 is a duplicate of D, for the production of yellow cotyledons, and is identical with WOODWORTH’S factor I , the only reason for the change being the use of I in designating a restriction factor.

The action of R, in this cross was the same as has been previously described. The dark hilum was easily distinguished from light hilum on tawny-pubescent plants, but the intensity of pigment was reduced on gray-pubescent plants, making it difficult to detect the presence of the restriction factor. In fact, all the gray-pubescent plants bore seed so much lighter in color than the tawny-pubescent plants that all were classified as having light hilums. Therefore only 3/16 of the total number of plants should come in the dark hilum class, and a 13:3 ratio should be expected.

The seed-coat color due to plastid pigment was perhaps the most variable character. The Fl seed coats produced a green color but it was somewhat intermediate. Fz results were in very close agreement with a 3: 1 ratio but there were all gradations in intensity of green. The pod color was no doubt responsible for much of this variation because light-colored pods were always associated with a “faded-out green.’’ Perhaps this can be explained as a result of the plants remaining in the field in the presence of bright sunlight for a considerable length of time after ripening. Black- podded plants always produce a normal green or yellow seed coat, and it seems possible that this can be accounted for by attributing to the black color the absorption of the sunlight which changes normal green seed


70 F. V. OWEN

coats to a yellowish green. In the light-colored pods the same effect was also produced on green cotyledons as described in Part I of this series.

There was a close approach to a 15: 1 ratio for cotyledon color in all Fz seed. The total numbers were 1686 with yellow cotyledons to 118 with green cotyledons, so the deviation is only 5.25 $6.9 from the expected numbers. Usually the seeds were easy to classify because the difference was clear-cut. Both of the parental varieties matured well in Wisconsin and no cotyledons were green due to immaturity. The F, plants also produced black pods, so there was no tendency for the cotyledon color to fade as it did on certain Fz plants with yellow pods.

The parental variety No. C4 has very dark brown or black pods. As has been stated the pigment responsible for this coloration has no relation to the anthocyanin or the brown phlobaphene-like pigment; yet very frequently, on black-podded plants, much of this pigment finds its way into the seed coat and produces a smudgy-brown coloration. This condition has often masked other pigments to such an extent that the classification of seed-coat colors has been very difficult. Fortunately the F2’s were grown on soil which brought out considerable mottling, but it was frequently necessary to treat the seed coats with acid in order to he sure of the classification. This test proved a great aid because a small amount of anthocyanin can be detected by the red color produced in acid solution.

TABLE 2 F ? progenies of JlandarinXGreen 1Yo. CJ.

__________~___ Green z~ersus Yellow seed-coat Tawny versus Gray pubescence Anthocyanin versus Brown pigment Black versus Yellow pods Purple versus White flowers Light versus Dark hilum

(calculated on basis of 13 : 3 ) Yellow versus Green cotyledons

(corrected figures)

OBSERVED

NUMB ER 8

252:85 262: 7.5 251:86 238 : 99 251 : 86 265: 72

321:16 315.9:21.1

DEVlATlON

0.75 9.25 1.75

14.75 1.75 8.5

5.1

DEV.

P. E. __

0.14 1 .73 0.33 2.75 0.33 1 . 8 2

1 .70

ACTU.4L RATIO

__- 2.96: 1 3.49:l 2.92: 1 2.40: 1 2.92:l

11.04:3

20.06: 1

Table 2 gives the observed data for each of the contrasted characters of 337 Fz plants. There is good agreement with expectation among all seven of these contrasted characters when considered separately except black versus yellow pods, This discrepancy perhaps can be accounted for by the fact that the classification was made from a few pods after thresh-


9: 3: 3 : l 9: 3: 3 : l 9: 3: 3: l 9: 3: 3 : l

39: 13: 9:32

39:13: 9:3 9: 3: 4:

39:13: 9:32 39:13: 9:3

9: 3: 3 : l 9: 3: 3 : l 9: 3: 3 : l

9: 3: 3:l 9: 3: 3 : l

9: 3: 3 : l

ing. Since the character is somewhat variable, further tests would be necessary to prove that there is any other reason for the results not being in accord with a 3 : 1 ratio.

In order to determine whether or not there was any linkage, each con- trasting pair of characters has been considered separately with every other pair. Table 3 gives these data in a condensed form. Twenty-one combinations are possible and the actual ratios as well as the theoretical are given in the table. The values of x2 and P are also included.

TABLE 3 FZ progenies of Green No. C4XMandarin-total qf 337 plants.

8.7: 3.2: 3.2:0.9 9.1: 2.9: 3.3:0.7 9.3: 2.7: 2.6:1.4 8.7: 3.3: 2.6:1.4

37.6: 12.7: 10.3:3.4

36.8:13.4:10.8:2.8

36.6:13.6:11.0:2.7 35.5:14.8: 9.7:4.0

8.4: 2.9: 3.5:1.2 8.7: 3.8: 2.6:0.9 8.3: 3.6: 3 .0 : l . l

8.6: 3.3: 3.3:O.S 9.4: 3.0: 2.5:l.O

9.3: 3.2: 2.6:0.9

9.0: 3.4: 3.6:

DIDz versus d& and W versus w and T versus t and RI versus rl and L versusl and I h versus I i and G versus g

'L

U

( L

U

U

G versus g and W versus w and T versus t and RI versus r1

and L versus1 and I h versus I'

U

U

U

U

I h versus I i and W versus w and T versus t and RI versus rl and L versus1

U

U

U

L versus 1 and W versus w and T versus t and RI versus rl

U

U

R1 versus rl and W versus w ' I and T versus t

T versus t and W versus w

l'BEOEETICAL RATIO

RITE INDEPENDENCE

45:15: 3 : l 45:15: 3:l 45:15: 3 : l 45:15: 3: l

195:45: 13:32 45:15: 3:l

ACTUAL RATIO

44.3 : 15.7 : 3.5 : 0,s 46.2:13.8: 3.8:0.2 44.7: 15.3: 3.0: 1 .O 41.9: 18.1: 3.5:O. 5

192.2 : 51.7: 9 .1 : 3 .O 44.1:15.9: 4.0:O.O

X2

2.07 4.68 0.05 6.45 1.83 7.44

1.21 2.69 4.54 5 . 5 8 1.52

2 . 7 1 2.25 3.50 4.95

3.48 5.30 3.89

2.17 2.07

1.41 -

P

0.56 0.20 0.9+ 0.09 0.61 0.06

0.75 0.45 0.21 0.14 0.68

0.44 0.33 0.33 0.18

0.33 0.15 0.28

0.54' 0.56

0.71

__.

-- 1 The observed ratio was corrected for a 15: 1 ratio for cotyledon color in making compari-

sons with the calculated figures. This is legitimate and advisable because no effort was made to grow these seeds to maturity in exactly that proportion.

These odd ratios are due to the fact that all gray-pubescent plants were classified as having light hilums. Really of these plants had dark hilums but they could not be readily identified.

It will be noted that five combinations give values of P less than 0.2 but four of these are combinations with black versus yellow pods where GENETICS 13: Ja 1928

72 F. 17. OWES

the original ratio was 2.4: 1; rather poor fits1 with other pairs of characters, therefore, would naturally result. One combination alone (plastid color in the seed coat versus cotyledon color) appears to be a case of linkage. In this case the value of P is 0.06 which means that the odds are only about 16:l against such a fit being due to chance but an examination of the actual results indicates that the odds may be greater. This linkage relation has been discussed in Part I of this series.

LINKAGES

Rz, factor for black pigment in seed coat (Per- I haps identical with WOODWORTH’S H Completely linked.

T, factor for tawny pubescence. i 1 1 Since the expression of E is somewhat quanti- tative the linkage could not be accurately measured, but there is proof for only a small percentage of crossing over.

Linkage appears to be almost complete with limited data.

T , factor for tawny pubescence. E, factor for early maturity.

PI, factor for glabrousness. RI, factor for black pigment in seed coat.

PI, factor for glabrousness. M , factor for black mottling

About 18 percent crossing over. (NAGAI and SAITO 1923)

G, factor for green seed coat, designated as V by) About 13 percent crossing over. (WOODWORTH

D1, factor for yellow cotyledon color 1921. Also the present study.) 1’ WOOD WORTH

Three linkage groups are necessary to explain the above linkages, but the factors cannot be arranged according to their respective positions. These groups are as follows:

RP-T-E Pi-RI-M

G-D1

LIST OF GENETIC FACTORS

The genetic factors used to explain the characters which have been studied in soybeans are as follows:

T, factor for tawny pubescence. Also increases the intensity of color in the seed coat; t , gray pubescence and a lighter seed-coat color. (WOODWORTH 1921) (Limited evidence indicates that there may be another factor which produces the same results as T but is independently inherited.)

W , factor for purple flower and hypocotyl color and also a complement of r2 for the production of black pigment in the seed coat; w, white flowers and green hypocotyls. (WOODWORTH 1923.)

1 A preferable method of calculating x2 as a measure of linkage follows this article.


R,, factor for black pigment in the seed-coat but without T the color is imperfect black; r,, brown pigment but without T the color is light brown or buff. (NAGAI 1921.)

Rz, also a factor for black pigment in seed coat, but entirely linked with T (possibly this factor is identical with WOODWORTH’S factor H ) ; r2, brown pigment (but is a complement of W . With rzr2 and W present, the pigment is imperfect black in appearance); rz’, brown pigment but not complementary to W.

B, found to be a complementary factor to Woodworth’s H for the production of black pigment in the seed-coat. (WOODWORTH 1921.)

0, factor for dull brown seed-coat color; 0 , reddish brown. (NAGAI 1921.) I, factor which inhibits pigment formation in the seed-coat-an allelomorphic series. i, has

no effect. I k , inhibits pigment to form an eyebrown pattern (identical with NACAI and SAITO’S,

1923, factor K ) . I:, inhibits pigment to the hilum-dominant to Ik and i. (Possibly the same as the factor

Z which NAGAI, 1921, describes as nearly completely inhibiting the formation of chromogenic substance in the seed-coat.)

I”, inhibits pigments entirely except that which may be due to mottling-dominant to Id , ZE, and i. (Identical with NACAI and SAITO’S factor H.)

M, factor for the production of black mottling upon a self-brown seed-coat; m, no effect. (NAGAI and SAITO 1923.)

G , factor for green seed-coat; g, yellow seed coat due to the green plastid color fading out a t maturity. (NAGAI, 1921, also WOODWORTH, 1921, but V was used for the symbol instead of G.)

D,, factor for yellow cotyledon color; d,, no effect. (WOODWORTH 1921.) D2, duplicate of 4, for causing the green cotyledon color to fade out a t maturity (identical

with WOODWORTH’S factor I ) ; dz, no effect. N , factor for normal hilum; n, abnormal hilum as in the Soysota variety. PI, factor for glabrousness; p1, pubescence. (NAGAI and SAITO 1923.) Pt, factor for pubescence, p2, glabrousness (relationship to PI and pl not known as yet except

that the order of dominance is reversed)-STEWART and WENTZ (1926). L, factor for black pods but also causes a black coloration of the seed-coat in certain instances;

1, brown pods. (WOODWORTH 1923.) S, factor for tall plant; s, short plant. (WOODWORTH, 1923, also reported by the writer in

table 8.) E, factor for early maturity; e, late maturity.

I n the above list of factors reference is made to all original descriptions of previous investigators. Those factors which do not have such a reference are described here for the first time. Of course, there is no definite proof that the factors which are considered synonymous are really identical, but it seems advisable to consider them as such until there is proof for the contrary.

SUMMARY

In all the crosses that have been made black seed coat has been dominant to brown but in one cross the dominance was incomplete. Results also indicate that different factors may produce black pigment; and a genetic interpretation has been given to account for the presence of RI GENETICS 13: Ja 1928

74 F. v. OWEN

and Rz, each of which is capable of producing anthocyanin pigment in the seed coat.

In one particular cross, the factor W , for purple flowers. was found to be a complement of 7 2 for the production of black pigment in the seed coat; but in all other cases purple-Aower color was independent of every other character.

These are dominant factors but their expression is very sensitive to the conditions under which the plant is grown. An environment unfavorable for the production of mottling is necessary for pigment to be completely restricted ; otherwise there may be considerable mottling in spite of the presence of restriction factors.

The factor T , for tawny pubescence, has proved to be very important in increasing the intensity of pigment. This is perhaps due to an actual increase in the amount of pigment.

In black-podded varieties a certain amount of the brownish-black oxidation pigment of the pod was sometimes evident in the seed coat.

Five linkages are postulated. The crossover percentage between G, the factor for green seed coat, and D, a factor for yellow cotyledons, was approximately 13 percent, the same as WOODWORTH (1921) has reported. The factor T , for tawny pubescence was completely linked to R,. All other linkages are quite tentative until more data are secured.

A complete list of all the genetic factors studied in soybeans has been given as well as those directly concerned with seed-coat colors.

Three restriction factors have been described.

LITERATURE CITED

BLAKESLEL, A. F., and AVERY, B. T., 1917 Adzuki beans and jimson weeds. Jour. Heredity 8: 125-131.

COMBES, R., 1912 Formation de pigments anthocyanique dCtermin6e dans les feuilles par la decortication annulaire des tiges. Ann. Sci. Nat. (Bot.) Paris, Ser. 9, Vol. 16: 1-53.

HOLLOWELL, E. A., 1924 Factors influencing the mottling of the soybean seed-coat. (Unpublished master’s thesis. Copy on file, Iowa State College Library, Ames, Iowa.)

MATSUMOTO, T., and TOMOYASU, R., 1925 Studies of purple speck of soybean seed. Ann. Phyt. Soc. of Japan 1: 1-14.

NAGAI, I., 1921 A genetico-physiological study on the formation of anthocyanin and brown pigments in plants. Jour. Coll. Agric., Imp. Univ., Tokyo 8: 1-92.

Japan. Jour. Bot. 2: 63-70. XAGAI, I., and SAITO, S., 1923 Linked factors in soybean. Japan. Jour. Bot 1: 121-136. OWEN, F. V., Hereditary and environmental factors which produce mottling in soybeans. (In

PIPER, C. V., and MORSE, W. J., 1923 The soybean. 329 pp. New York: McCraw-Hill Book C O . STEWART, RALPH T., and WENTZ, JOHN B., 1926 .I recessive glabrous character in soybeans.

1924 Observations on the somatic segregation in soybeans.

Jour. Agric. Res. 34: 559-587.

Jour. Amer. Soc. Agron. 18: 997-1009.

SEED-COAT COLOR I N SOYBEA4NS 75

TAKAHASHI, Y., and FUKUYAMA, J., 1919 Morphological and genetic studies on the soybean. (In Japanese) Hokkaido Agric. Exp. Sta. Rpt. No. 10, 100 pp., Abs. in Jour. Coll. Agric., Imp. Univ., Tokyo 8: 91-92 (1921).

TERAO, H., 1918 Maternal inheritance in the soybean. Amer. Nat. 52: 51-56. WOODWORTH, C. M., 1921 Inheritance of cotyledon, seed-coat, hilum, and pubescence colors

1923 Inheritance of growth habit, pod color, and Bower color in soybeans. Jour. Amer.

WOODWORTH, C. M., and COLE, L. J., 1924 Mottling of soybeans. Jour. Heredity 15: 349-354.

in soybeans. Genetics 6 : 487-553.

Soc. Agron. 15: 481495.

APPENDIX

TABLE 4 FP progenies of b r o w (rli) X yellow with black hilum (RI In).

Soysota (Brown No. 30) XManchu (No. 59)

Calculated

Difference

Brown (No. 12)XManchu (No. 1)

Calculated

Difference

Brown (No. 12)XManchu (No 16) Calculated

Difference

Total of Fe progenies Calculated

Difference

PROQENY NUMBERS

213 224 217

Total

186 190

Total

8

DARK KlLUY

Black RIIi

58 53 84

195 192.4

+2.6 __

71 128 199 181.7

+17.3

190 185.6

$4.4

584 559.7

--

4-24.3

Brown r,I'

14 13 28 55 64.1

-9.1

11 41 52 60.6 -- -8.6

62 61.9

+o. 1

169 186.6

-17.6

--

___

SELF-COLORED

SEED-COAT

Black Rd

18 13 33 64 64. I

-0.1 __

18 38 56 60.6 -

-4.6

51 61 .9

-10.9

171 186.6

-_ -15.6

___ Brown

r i --

9 6

13 28 21.4

+6.6 __

5 11 16 20.2

-4.2 -

27 20.6

+6.4 -

71 62.2

+8.8 -

-- x2=2.53 P=0.47

-- xf=4.09 P=0.25

-- xz =4.95 P=0.18

- x2= 5.26 P=0.16


76

Mottled with brown

40 37.1

f 2 . 9

10

11.3 -1.3

~ - _ _ _

--_I_

68 68

I;. v. O W E S

Self-brown

lo 12.4

-2.4

2

3.7 f 1 . 7

26 22.7

+ 3 . 3

TABLE 5 F2 progenies of yellow (tIh)Xbrozan seed coat (Ti) .

19 21.4

- 2 . 4 ~ _ _ _

137 137.8 -0 .8

MandarinX Brown (No. 12) Calculated Dzerence

10 7.1

f 2 . 9

48 45.9

+2.1

MandarinX Soysota (Brown No. 30) Calculated Difference

Yellow (No. C2) X Soysota (Brown No. 30) Calculated Difference

AksarbenXBrown (No. 60 Calculated D8erence

Total observed Calculated Difference

'ROGEW

VmdBER

-

197

231

208

164a

TAWNY PUBESCENCE a u y PUBESCENCE

Mottled vith brown

10 12.4

- 2 . 4

5

3.7 +1.3

18 22.7

-4 .7

5 7 .1

- 2 . 1

38 45.9

-7.9

Self-brown

6 4.1

f 1 . 9

3

I .3 4 - 1 . 7

--

9 7.6

f 1 . 4

4 2 .J

+ 1 . 6

22 15.3

4-6.7

xz = 2.04 P = 0 . 5 7

x2 = 3.61

P = 0 . 3 1

x2=1.71 P=0.65

x2=3.14 P=0.37

x2= 4.39 P = 0 . 2 3

SEED-COAT COLOR IN SOYBEANS

Fa Progenies from seeds with light black hilum Deviation from calculated

Deviation from calculated



--.

77

Fa Progenies from seeds with black dark hilum Deviation from calculated

F3 Progenies from seeds with brown light hilum Deviation from calculated

_.

F3 Progenies from seeds with dark brown hilum

TABLE 6 Dark black hilum (RIZi) Xlight brown hilum (rlZh).

1

2

3

2

2

NUMBER

OR PROW. NIES

I- F* Progeny No. 1 (Manchu No. 1XYellow

No. 12)

l-

LIGHT HILUM

Black pigment

120

124.4 -4.4

45 +2.8 36

-1.5 49

-1.3 17

-1 57

---

Brown pigment

36

41.4 -5.4

12 -2.1

18 +1.3

--

---

71 f 1 . 8 38

DARK HILUM

Black pigment

51

41.4 +9.6

16 +1.9

14 +1.5

7 +1

11 +1.2 76

--

Brown pigment

14

13.8 +0.2

-I_

2 -2.7

---

2 -1.3

-_--

26 -1 .7

44

x2=3.10

P=0 .38


78

_-_ Brown

Pigment

+0.9

TABLE 7 Tawny-pubescent, Dark black hilzm (IsR1) X Gray-pitbesce?zt, Light brown hilum (tIhrl).

_-__ Black Brown

Pigment Pigment _~_______.-

3 5 6 -3.4 +3.2 x2=5.75

P=0.33

Green (No. C4) X.4ksarben Deviation from calculated

Brown Pigment --__

-0.3

__

-1.2

--_

+1.4

-__- 62

68 614.7

~

47 14 61

+3.1

-- SO

+2.6

_____ 195 174.8 t20 .2

Yellow (No. C2) X Manchu (No. 1) Deviation from calculated

~

Black Pigmer

6 5 -1.

__-

4 6 +o., __

6 9 +4.*

40 6 4

44 -9..

_-- 48

7 55

-2.'

58 + lo . (

177 174.q +2.

MandarinX Green (No. C1) Deviation from calculated

4 +2.3

+0.5

MandarinXManchu (No. 1 )


11 2 f 4 . 1 -0.3 x2=8.00

P=0.16 - ~ - _ _ -

2 7 2 $0.9 -0.1 x2=8.33

P=0.14

Manchu (No. 59) X Mandarin

14

15 -2.8


____ 51

1 5 56

-15.1

Green (No. C4) X Mandarin Deviation from calculated

20 1

21 -2.7

Calculated on basis of 27: 9: 9: 3: 12: 4 ratio Difference

x2=11.06 P=O.O5

31

+0.7

269

P=0.57 +2.7 $8.2 x*=3.84

342

14 53 22 -1.8 -10.2 +0.9

300 292

Total

__ 282 272

Total

x2=4.44 P=0.49

13

- Total 58

58.3 -0.3

TAWNY PUBESCENCE QRAY PUBESCENCE __--

Light Hilum 1 Dark Hilum

212 87 233.1 77.7

-21.1 +9.3 x2=5.60 P=0.35 -

Black Pigmeni

20 +1.C

10 -5.t

7 -6,s

- 161 14

175 +15.1

-- 134 28

162 -ll.E

140 -2.2

_- 514 524.5

-10.:

-__~--l--- I----

-1-1-1

SEED-COAT COLOR I N SOYBEANS

- Observed Calculated Deviation

79

SGR, SGr, -- 85 29 8 5 . 2 28.4 - . 2 + . 6

TABLE 8

Showing independence between S, the dominant factor for height of plant and RI and G-Progeny NO. 319-No. 57 (Srig) X NO. Ci (SRiG).

Nril; __.-

8 5 13.9

-5 .9

N T I ~

4 . 6 -0.4

24 9 28.4 1 9 . 5

-4 .4 - . 5

N R J i -- 41 41.8

-0 .8

x2=1.95 :.5 1 ‘ i . 5 I 3 . 2 1 P=0.96

f 4 . 6 -1 .5 +1.5 - . 2

NRti

11 13.9

-2 .1

TABLE 9

Manchu ( R J i N ) XSoysota (rlin)

nRJi

17 13.9

+3.1

Observed Calculated Difference

nRd nrJi nrd

7 6 4 x2=9.70 4 . 6 4 . 6 1.5 P=O.22

_I--- -

+ 2 . 4 +1.4 +2.5

RsN Rm rrY r m

Observed 49 23 21 7 Calculated 56.25 18.75 18.75 6.25 Difference -7.25 +4.25 $2.25 +0.75

x*=2.26 P=O.50


Documents

INHERITANCE STUDIES IN SOYBEANS. 111. SEED-COAT ...INHERITANCE STUDIES IN SOYBEANS. 111. SEED-COAT COLOR AND SUMMARY OF ALL OTHER MENDELIAN CHARACTERS THUS FAR REPORTED' F. V. OWEN